English

Semantic-Enhanced Explainable Finetuning for Open-Domain Dialogues

Computation and Language 2022-05-25 v2

Abstract

This paper propose to combine pretrained language models with the modular dialogue paradigm for open-domain dialogue modeling. Our method, semantic-enhanced finetuning, instantiates conversation understanding, planning, and response generation as a language model finetuning task. At inference, we disentangle semantic and token variations by specifying sampling methods and constraints for each module separately. For training and evaluation, we present X-Weibo, a Chinese multi-turn open-domain dialogue dataset with automatic annotation for emotions, DAs, and topical words. Experiments show that semantic-enhanced finetuning outperforms strong baselines on non-semantic and semantic metrics, improves the human-evaluated relevance, coherence, and informativeness, and exhibits considerable controllability over semantic variables.

Keywords

Cite

@article{arxiv.2106.03065,
  title  = {Semantic-Enhanced Explainable Finetuning for Open-Domain Dialogues},
  author = {Yinhe Zheng and Yida Wang and Pei Ke and Zhenyu Yang and Minlie Huang},
  journal= {arXiv preprint arXiv:2106.03065},
  year   = {2022}
}

Comments

Under review

R2 v1 2026-06-24T02:52:44.994Z